Knostic

AgentMesh continuously discovers, tracks, and scans AI agent skills, MCP servers, and VS Code extensions for prompt injection and supply-chain threats.

AI agent security: Kirin applies least privilege to agent actions, monitors every tool call, and blocks unsafe behavior before it executes.

Kirin secures AI coding assistants by preventing prompt injections, rogue behavior, and unsafe execution.

MCP security posture management: inventory and validate every MCP server, enforce secure configuration, flag drift, and block rogue connectors.

The page explains how evasion, poisoning, and inference attacks compromise enterprise AI models and how Knostic mitigates them in real time.

Knostic's free agentic threat intel: 4 source-validated VS Code extension findings (May 13–20, 2026), including BCAI Rosetta with 13,480 installs on Cursor.

The company highlights the rapid adoption of coding agents, from platforms like rentahuman.ai to production environments, and outlines security measures for detection, monitoring, and control.

Implement attribute-based access control for AI assistants and agents with clear steps, PDP/PEPs, policies, RBAC migration, and audit-ready KPIs.

Emerging From Stealth, Knostic Named 2024 RSA Conference Launch Pad Finalist.

Compare Bright Security vs. Corridor across shadow AI governance, AI assistant security, prompt gateway firewalls, and AI coding safety.

CodeRelay: a coordinated campaign of 13 malicious VSIX packages across 12 VS Code extensions disguised as developer tools. Analysis, IOCs, and defenses.

Knostic serves as the control plane for agentic AI code stacks, with information about its team, investors, and research.

Knostic protects data, users, and AI tools.

Pick a time for a personalized Knostic product demo and get a calendar invite.

Knostic provides a solution brief on detecting and responding to LLM data leakage in AI chatbots, addressing data exposure and security risks.

Knostic secures Claude Code, Cowork, MCP servers, and skills to enable controlled adoption of AI agents in business environments.

Effective as of April 2024 (last reviewed April 2026)

GitHubRepository owner

Knostic secures **$16 million** in funding to develop AI access controls, enhancing data security for platforms like **Microsoft Copilot** and other generative AI systems. The company focuses on safeguarding sensitive data within AI workflows, addressing growing concerns over unauthorized data exposure in enterprise AI tools.

Jun 2025Virginia BusinessNews article

Knostic secures **$11 million** to develop **need-to-know access controls** for generative AI, targeting data leaks in enterprise LLMs. The funding expands its safety layer for tools like **Microsoft 365 Copilot** and **Glean**, supporting AI adoption while mitigating unauthorized data exposure. Total funding now reaches **$14 million**.

Mar 2025Pulse 2.0News article

Knostic raises $11 million to address AI-driven data leaks in enterprise search tools. The cybersecurity startup focuses on preventing oversharing of sensitive information in AI-powered business intelligence systems. With total funding now at $14 million, Knostic aims to strengthen security in enterprise AI applications.

Mar 2025Unite.AINews article